Nonparametric Empirical Bayes Estimation of the Matrix Parameter of the Wishart Distribution
โ Scribed by Marianna Pensky
- Publisher
- Elsevier Science
- Year
- 1999
- Tongue
- English
- Weight
- 168 KB
- Volume
- 69
- Category
- Article
- ISSN
- 0047-259X
No coin nor oath required. For personal study only.
โฆ Synopsis
We consider independent pairs (X 1 , 7 1 ), (X 2 , 7 2 ), ..., (X n , 7 n ), where each 7 i is distributed according to some unknown density function g(7) and, given 7 i =7, X i has conditional density function q(x | 7) of the Wishart type. In each pair the first component is observable but the second is not. After the (n+1) th observation X n+1 is obtained, the objective is to estimate 7 n+1 corresponding to X n+1 . This estimator is called the empirical Bayes (EB) estimator of 7. An EB estimator of 7 is constructed without any parametric assumptions on g(7). Its posterior mean square risk is examined, and the estimator is demonstrated to be pointwise asymptotically optimal.
๐ SIMILAR VOLUMES
A mean-squared error comparison of smooth empirical Bayes and Bayes estimators for the Weibull and gamma scale parameters is studied based on a computer simulation. The smooth empirical Bayes estimators are determined as functions of up to 15 past estimates of the parameter of interest. Results indi
## The parameters of the function f(t) =c(e-& -e -\* l ) are related in a simple way to the moments I t Y ( t ) dt (n =0, 1 , 2 ) . Using empirical values of I, the moments can be estimated by numerical -0 integration. Therefrom estimates of the parameters are obtained by elementary algebra.